These Tiny, Simple Robots Can Work In Group And Navigate Tricky Terrains

Big, complex robots like the Atlas or Spot with agility indicate the effort and hope for the future of robotics, but that doesn't mean we can look down on tiny, simple robots when they work as an adaptable group. And these "tribots," created based on the model of ants, can overcome obstacles through teamwork.

The team behind these tribots include researchers from Osaka University and EPFL. While these simple, tiny, light robots move like inchworms more than ants, they can fling up and forward when needed. The research team modeled how these bots work on trap-jaw ants, which crawl and jump, with different roles such as leader, worker, and explorer. Though each robot is not really intelligent, working as a collective helps them deploy their abilities intelligently.

Considering that, a team of tribots will be able to cross a complex terrain. For this to be done, an explorer moves ahead, senses obstacles, and relays its locations and dimensions to the other robots in the team. Next, the leader assigns units of workers to push the obstacles out of the way. If this first attempt doesn't work, an explorer will try hopping over the obstacle; in this case, if it succeeds, it can transfer the information to the team so that they can follow.

For now, the process still happens slowly, as in the video below, which illustrated the action at 16x speed. However, the point here is not rapidity, but more about the simplicity and adaptability of deployment.

Each bot weighs only 10 grams. Additionally, they are essentially PCBs attached with some grip points and mechanical bits, which means we can easily mass-produce them, said the paper which was published in the journal Nature. And if they are cheap to make, we can deploy dozens or even hundreds of them on a target terrain so that they can characterize it, take measurements, or search for heat hot spots or radiation, etc., within a few hours.

If the bots moved faster, we can use them in a dining room or kitchen to scoot plates or collect crumbs.

The advantage of swarm-based robots is that in case something goes wrong, they do not fail catastrophically, the collective would still persist and we can easily replace a failed individual bot.

Co-designer Jamie Paik from EPFL said: 

While these bots are just prototypes, they have already shown a major advance over other collective systems of bots.