Posts

This article says animals in groups, like bees, fish, or birds, can solve problems together through many small local interactions, and engineers are using those same ideas to build robot swarms that can act without a central controller.

This paper describes an improved ant-inspired algorithm for finding good paths, such as routes for robots or navigation systems. Instead of searching too randomly, it puts more attention on promising areas early, strengthens better routes as it learns, and discourages unnecessary turns, so it finds smoother and better paths faster.

New tools such as AI, automated tracking, drones, and even virtual reality are letting scientists watch animal groups like flocks, schools, and swarms in far more detail than before.

This paper describes a new way for groups of robots to make a shared choice, based on how ants seem to decide on the best new nest.

This paper looks at how robot swarms can move more smoothly by not only reacting to what nearby robots are doing now, but also by briefly predicting where they are about to move next.

8-10 June 2026 in sunny Germany. ANTS solicits contributions dealing with any aspect of swarm intelligence. Typical, but not exclusive, topics of interest are:

Do emotions and varying environmental stimuli affect the swarms choice of location for the hive? https://arxiv.org/abs/2603.09963

If you’re the kind of person who can’t help finishing other peoples’ sentences for them, don’t worry; you can blame it on evolution.

Here are 13 peer-reviewed publications on the infrastructure to make embodied intelligence possible. In order you will find: 1. Human–robot object handover This review surveys how robots pass objects to people and receive them back, framing handover as a foundational capability for human–robot collaboration.

New website is live! In August 2024 the former adappt.io was renamed to adaptive-emergent.com to narrow the scope from web3 distributed networks to biomimetic ones (read the three part series on this starting here).